It is a strange feeling to come
to the end of the major part of the woodworking after three and a half
years. There are still things like the gunwale rubber and drink
holders, and the like, but the constructional part of this boat is
finished. Two of my least favoured jobs remain to be done: the deck
glue-down and the varnishing. There are still a few minor electrical
tasks to be completed too, and the windscreens to fit, but right now I can stand on
the back deck, looking over the cockpit and foredeck, and
marvel at this thing which has emerged out of the chaos which surrounds
it.
Anyway,
on to the conclusion. The
planking is complete except for the outer edges of the hatch covers,
where there has to be a bit of wood removed to allow them to open.
While I was cutting these parts a strange visitor appeared out of the
fence in my table saw. It apparently lives there, and seemed quite at
home covered in sawdust. If I knew what it was called I might even
rename the boat after it. But I don't.
Over the hatch covers there is a small drop in height from the rest of the deck, about 1 mm., because the hatches are not quite thick enough to ride flush; but there is no need to correct this. The hatches will have metal trim around them, and the trim will level them to the deck. Having a potential drop will merely serve to ensure that the trim makes a firm closure.
The job of gluing down the decking starts at the king plank in front of the windscreen. Some assistance is available here in the form of the spotlight. It is mounted on a plinth, and, by tightening its handle under the dash, downward pressure is exerted which clamps down on the king plank. Downward pressure on the light's post is designed to point the light itself down, so that is resisted by a wedged block. The rest of the plank is fixed by clamps at hatch opening level, and by the ply washers along the sides, as are some neighbouring planks too.
The spotlight on its plinth.
The king plank and three neighbouring port side strips are glued down.
A 3 mm. shim stick checks the gaps between planks (left), and after the glue has dried (right).
Over the hatch covers there is a small drop in height from the rest of the deck, about 1 mm., because the hatches are not quite thick enough to ride flush; but there is no need to correct this. The hatches will have metal trim around them, and the trim will level them to the deck. Having a potential drop will merely serve to ensure that the trim makes a firm closure.
The job of gluing down the decking starts at the king plank in front of the windscreen. Some assistance is available here in the form of the spotlight. It is mounted on a plinth, and, by tightening its handle under the dash, downward pressure is exerted which clamps down on the king plank. Downward pressure on the light's post is designed to point the light itself down, so that is resisted by a wedged block. The rest of the plank is fixed by clamps at hatch opening level, and by the ply washers along the sides, as are some neighbouring planks too.
The spotlight on its plinth.
The king plank and three neighbouring port side strips are glued down.
When the
short planks behind the hatch are completed you come up against the
boards which run full length from the windscreen to the covering board.
It is necessary to have the covering boards glued on before attempting
these longer planks, just to make sure that their forward alignment is
correct. If they shift to leave too small a gap between themselves and
the covering boards that it not too much of a problem, but if they make
too large a gap it is. You might think that having all these washers
and screws in would hold the planks in their correct position, but in
fact there is quite a bit of play in them, and they have to be adjusted
with 3 mm. shim sticks all the time during glue-down to ensure uniform
gaps.
A 3 mm. shim stick checks the gaps between planks (left), and after the glue has dried (right).
The
glue-down
of the covering boards is a clamp hungry manoeuvre which
takes a bit of improvisation too. The original plan for this boat did
not require a lower rubbing strake to be fitted, but I have found it
invaluable as a seat for the clamps. Without it I would have had to
have used staples, with all the unsightly scars they would have left.
The port covering board is glued down.
The port covering board is glued down.
While
the
covering board is setting I can get on with the starboard planking
behind the hatch. Three short strips are glued down there. This
completes the section between the windscreen and the hatch.
While
waiting
for this lot of epoxy to set I have been sanding back some of
the port covering board to see what the glue lines will look like. The
pre-sanded and post-sanded photos are not from the same board, but the
effect is very similar. One is the starboard and the other is the port
covering board. At least without finish on them they look pretty good.
Whether that stands up to a varnish time will tell.
Before (left), and after (right) sanding the forward covering board joints.
Before (left), and after (right) sanding the forward covering board joints.
Attaching
the
starboard covering board is basically the same as the port, except
for the treatment of the bow. Here, there is a lot of resistance to the
two halves meeting at a tight junction. I racked my brains trying to
come up with some sort of wedged clamp system which could be used to
hold the two halves together, but then it occurred to me that the sheer
is going to be used to anchor the gunwale, so why not some clamps as
well?
Three simple blocks are screwed through the hull side into the sheer where the hole will later be covered by the gunwale rubber, after filling, of course. These push up tightly against the tip of the starboard covering board and force it against the port, which is already glued down. Some downward pressure on the joint line further forces the two together, and a fairly tight glue line results. One of those traditional bow ornaments would also have been very useful, as it could have covered any screws which might have been called into service for this task.
The starboard covering board is glued down, and three simple clamps on the sheer force a tight joint.
Three simple blocks are screwed through the hull side into the sheer where the hole will later be covered by the gunwale rubber, after filling, of course. These push up tightly against the tip of the starboard covering board and force it against the port, which is already glued down. Some downward pressure on the joint line further forces the two together, and a fairly tight glue line results. One of those traditional bow ornaments would also have been very useful, as it could have covered any screws which might have been called into service for this task.
The starboard covering board is glued down, and three simple clamps on the sheer force a tight joint.
There
is
still a lot of gluing to be done on this deck, but it is beginning
to take shape, and can be seen in its almost ready state now, from the
port.
Next
come
the king plank between the bow and the hatch, and the three planks
on either side of it. I have been giving some thought to the
termination of the decking planks at the hatch. As it stands they
merely end abruptly, so that the caulking will have to end there too.
It would be neater if there were some sort of dam wall to hold the
caulking in, and up to which the decking planks could run. It needs
only be a narrow strip of 5 mm. mahogany running athwart to the planks.
To allow for it I have either to advance the king plank and the
accompanying
members forwards by the width of the strip, or take the router to the
hatch end after they have been glued down. That will probably result in
a neater cut, so on with the glue-down.
All the strips in front of the hatch are now glued down.
Both sides glued down, then washers removed.
Finally,
the
side strips are routed and glued on.
I would have to say that the router method of creating a groove between the covering boards and the decking planks was problematic. It went very smoothly on the rear deck, but on the foredeck the camber at the edge is greater, so that there is no flat surface for the router base to ride on. Also, at the sharpest part of the curve on the hull there is hardly any of the router fence in contact with the hull. The result is a somewhat wavy groove which will not look so good filled with caulking. The worst part of it occurs near the bow.
A wavy groove.
All the strips in front of the hatch are now glued down.
Once
the
third strip from the king plank is reached it is time to pause. The
next one has to align precisely with the side of the hatch, so the gap
between the third and fourth planks is critical. If it is too big, some
of the inner edge of plank four can be shaved off where it is rebated
for the hatch. If it is too small, some of its inner edge where it is
not rebated will have to go. So I have to wait for the glue to dry
before proceeding.
Here the hatch-side planks are lifted away from
the central, completed part of the deck.
Here the hatch-side planks are lifted away from
the central, completed part of the deck.
As it happened the spacing was
right, so the next planks went down without too much fuss. Working on
the side decking here I can only go as far as the second last plank.
The last one is not cut to fit the covering board (or, rather, it
presently fits it tightly) so will have to be routed. But the router
will not be able to reach its aft end, next to the windscreen. That end
will have to be finished by hand in line with the routed edge further
forward on the plank. Therefore, the plank has to be removable. The
routing will be done with it stuck down on double sided tape. That
means that the epoxy which has oozed out onto the subdeck where the
last plank is to go will have to be dried first.
The port side of the foredeck is finished except for the last plank.
Epoxy ooze is a constant bother, and continues for hours after it is first laid. The drips and runs seen here are on the sides of the hatch opening, and have been wiped away three times already, only to reappear.
Runs on the hatch opening.
The port side of the foredeck is finished except for the last plank.
Epoxy ooze is a constant bother, and continues for hours after it is first laid. The drips and runs seen here are on the sides of the hatch opening, and have been wiped away three times already, only to reappear.
Runs on the hatch opening.
A full day of gluing
and screwing is needed to do both sides, minus the outer planks. I
eventually gave up using masking tapes as the epoxy gets everywhere
anyway. It is best just to wipe it up with acetone or metho and sand
the wood later, which was going to happen anyway. The dirtiest job is
going back after the epoxy has partially gone off, and scraping out the
excess from the caulking gaps, so that there will be enough room to
squeeze in the Sikaflex. With all that coming out to the surface it is
then a good idea to prevent the ply washers from sticking to the
decking by inserting silicone paper strips between the two. It is quite
tedious. The thought that I am nearly done with the epoxy for good
delights me.
Both sides glued down, then washers removed.
I would have to say that the router method of creating a groove between the covering boards and the decking planks was problematic. It went very smoothly on the rear deck, but on the foredeck the camber at the edge is greater, so that there is no flat surface for the router base to ride on. Also, at the sharpest part of the curve on the hull there is hardly any of the router fence in contact with the hull. The result is a somewhat wavy groove which will not look so good filled with caulking. The worst part of it occurs near the bow.
A wavy groove.
Because
the
decking planks at this level are presenting almost total end grain
it is not easy to smooth out the groove with hand sanding either. The
alternative method of creating this groove would have been to hand
shape the planks to the covering board, and I think that that would
have resulted in a number of irregularities as well, so I don't really
know what to do differently next time. I have a sinking feeling,
however, that this is going to look a lot worse when the caulking goes
in.
Even if you manage to avoid these pitfalls, the router bit is running against the grain on the port side. It has, after all, to trim the decking planks, not the covering boards which are already shaped. So the bit has to run uphill against the grain on one side of the boat. With White Beech the effect is more like a cheese grater than a router, and there is quite a bit of feathered grain which need to be sanded off. The cut is cleaner on the mahogany, but unless the router bit is perfectly sharp there is a potential for it to run amok. It did so to the extent that it melted and snapped off in my case, and it wasn't all that blunt to begin with!
To ruin your good work after all this time, in an area which cannot be disguised, is deeply disappointing, to say the least. The neighbours may begin to complain about the language, as well as the noise. Never mind, if it looks really obvious I can cover it with some judiciously chosen piece of deck hardware. So, now there is only the hatch covers to complete before moving on the the gunwales.
Hatch covers drying.
They should have been simple, and, indeed, one of them was. On the starboard hatch, however, a depression had developed in the ply subdeck, which was quite noticeable, even with the decking planks attached. Perhaps it is where I sat more often than not when getting into the motor compartment. Whatever the cause, the outcome was a far from fair line on the decking. So, it had to be packed up somehow. I elected to do this with wooden shims between the subdeck and the deck, which created a space which I filled with epoxy thickened with West 410 Microlight, which is what I used to fair the hull all those aeons ago. It had to be applied thickly so as not to run out into the gaps, but even so, the difference between the good hatch and the bad one is obvious by the colour of the gaps. The bad one is full of pale Microlight, which has to be scraped out when it has thickened sufficiently.
The white looking Microlight seen in the caulking gaps on the starboard hatch (left), and clean lines on the port.
The method did result in a fair deck line, so the foredeck is now complete and ready for caulking.
Even if you manage to avoid these pitfalls, the router bit is running against the grain on the port side. It has, after all, to trim the decking planks, not the covering boards which are already shaped. So the bit has to run uphill against the grain on one side of the boat. With White Beech the effect is more like a cheese grater than a router, and there is quite a bit of feathered grain which need to be sanded off. The cut is cleaner on the mahogany, but unless the router bit is perfectly sharp there is a potential for it to run amok. It did so to the extent that it melted and snapped off in my case, and it wasn't all that blunt to begin with!
To ruin your good work after all this time, in an area which cannot be disguised, is deeply disappointing, to say the least. The neighbours may begin to complain about the language, as well as the noise. Never mind, if it looks really obvious I can cover it with some judiciously chosen piece of deck hardware. So, now there is only the hatch covers to complete before moving on the the gunwales.
Hatch covers drying.
They should have been simple, and, indeed, one of them was. On the starboard hatch, however, a depression had developed in the ply subdeck, which was quite noticeable, even with the decking planks attached. Perhaps it is where I sat more often than not when getting into the motor compartment. Whatever the cause, the outcome was a far from fair line on the decking. So, it had to be packed up somehow. I elected to do this with wooden shims between the subdeck and the deck, which created a space which I filled with epoxy thickened with West 410 Microlight, which is what I used to fair the hull all those aeons ago. It had to be applied thickly so as not to run out into the gaps, but even so, the difference between the good hatch and the bad one is obvious by the colour of the gaps. The bad one is full of pale Microlight, which has to be scraped out when it has thickened sufficiently.
The white looking Microlight seen in the caulking gaps on the starboard hatch (left), and clean lines on the port.
The method did result in a fair deck line, so the foredeck is now complete and ready for caulking.
Having
had
time to reflect on this method of decking, I can say that I think
it is slow, messy and imprecise. The certainty of being to able to
screw down a plank and then cap the screw hole has a lot to recommend
it, but is not practical in 5 mm. thick boards. The Gougeon Bros.
method of screw and washer fixation would be a lot better if one were
not screwing into a ply subdeck of only 6 mm., and is clearly designed
for heavier decking than mine. On the other hand, prefabricated plydeck
looks just that, and has the added disadvantage of having only 1 mm. of
the surface timber, usually teak and ash.
The Selway Fisher design, which recommends a second skin of ply, or narrow teak strips with epoxy caulking, is having a punt in both markets, so I guess that you trim your cloth to match your pocket, and your patience. I am hopeful that the finished appearance of this deck will justify the time taken, but in order not to ruin it altogether with the caulking I am going to apply a coat or two of sacrificial varnish before the caulking. This will then be sanded away, but will help to protect the Beech from black staining with the Sikaflex. Laying the deck is only just the beginning of the work!
The Selway Fisher design, which recommends a second skin of ply, or narrow teak strips with epoxy caulking, is having a punt in both markets, so I guess that you trim your cloth to match your pocket, and your patience. I am hopeful that the finished appearance of this deck will justify the time taken, but in order not to ruin it altogether with the caulking I am going to apply a coat or two of sacrificial varnish before the caulking. This will then be sanded away, but will help to protect the Beech from black staining with the Sikaflex. Laying the deck is only just the beginning of the work!
146. Gunwale Rubbers
The
Andrews designed Slipper Launches achieved gunwale protection by
attaching chromed brass rubbing strips to the edges of the covering
boards. This was possible because the covering boards were the real
thing, not veneers on a ply subdeck. The plans for Ariadne call for a
half round profile edging of hardwood, 36 mm. in diameter, which is
screwed into the side of the hull and the sheer strakes. That would
make it far too big to take a rubbing strip to cover it, and
it is obviously meant to replace such a device.
I have a problem with the semicircular profile, it being so dumb looking on an otherwise elegant craft. But, considering that the subdeck and deck make up a thickness of 11 mm., if the screws are not to enter the sheer strakes right on their edges there has to be a reasonably wide timber, assuming that they are to go through the rubber's centre. 12 gauge screws are specified, so even at 36 mm. diameter there will only be a gap of about 4 mm. between the screw hole and the top of the sheer. The plan makes no mention of gluing the rubber on to the hull and deck side, and it may be an attractive proposition not to, so that any damaged parts could be replaced at a later date. But with this fairly tenuous grip on the boat I think it needs gluing.
I am using the same Merbau timber that I used for the rubbing strakes already completed, and that means three lengths of wood scarfed together again. With 18 mm. of thickness and 36 mm. of width they are going to offer a lot of resistance to bending, especially at the stern where they have to follow the line of the deck in a downwards curve.
This time, in order to try to avoid the unyielding sections where the scarf joints are, I will scarf the short segment into the middle of the rubber. (The timber is 2.7 metres long, so only 2-1/2 lengths are needed). That way, the sharp downward curve at the rear of the gunwale will occur in an unscarfed section, and will hopefully follow the deck line more accurately than the lower rubbing strake did.
Fitting starts at the bow where the angle between the stem and the hull has to transferred to the rubber. With the rest of its length supported by a ceiling line, the first screw is driven into the sheer.
A line from the garage door holds the rubber while the bow is connected.
From then on it is merely a matter of moving the rubber down to deck level at each screw, and push it tightly against the hull while they are driven in. The first few are easy, but after that the increasing curvature of the hull makes it more and more difficult. Some help is found from bracing, against anything which will offer resistance. There is no possibility of clamping here as there was with the lower running strake until you get to within reach of the hatch opening. Then, however, the pull of the clamp tends to approximate the upper end of the rubber against the deck, but leave an opening at its lower edge. Again, braces can help close the gaps.
Extra help needed to clamp the rubber to the hull.
Clamping the repair.
But, the result is the same again. The snap is straight through the original, splitting apart the epoxy join. So much for its superior strength.
Top
of Page
I have a problem with the semicircular profile, it being so dumb looking on an otherwise elegant craft. But, considering that the subdeck and deck make up a thickness of 11 mm., if the screws are not to enter the sheer strakes right on their edges there has to be a reasonably wide timber, assuming that they are to go through the rubber's centre. 12 gauge screws are specified, so even at 36 mm. diameter there will only be a gap of about 4 mm. between the screw hole and the top of the sheer. The plan makes no mention of gluing the rubber on to the hull and deck side, and it may be an attractive proposition not to, so that any damaged parts could be replaced at a later date. But with this fairly tenuous grip on the boat I think it needs gluing.
I am using the same Merbau timber that I used for the rubbing strakes already completed, and that means three lengths of wood scarfed together again. With 18 mm. of thickness and 36 mm. of width they are going to offer a lot of resistance to bending, especially at the stern where they have to follow the line of the deck in a downwards curve.
This time, in order to try to avoid the unyielding sections where the scarf joints are, I will scarf the short segment into the middle of the rubber. (The timber is 2.7 metres long, so only 2-1/2 lengths are needed). That way, the sharp downward curve at the rear of the gunwale will occur in an unscarfed section, and will hopefully follow the deck line more accurately than the lower rubbing strake did.
Fitting starts at the bow where the angle between the stem and the hull has to transferred to the rubber. With the rest of its length supported by a ceiling line, the first screw is driven into the sheer.
A line from the garage door holds the rubber while the bow is connected.
From then on it is merely a matter of moving the rubber down to deck level at each screw, and push it tightly against the hull while they are driven in. The first few are easy, but after that the increasing curvature of the hull makes it more and more difficult. Some help is found from bracing, against anything which will offer resistance. There is no possibility of clamping here as there was with the lower running strake until you get to within reach of the hatch opening. Then, however, the pull of the clamp tends to approximate the upper end of the rubber against the deck, but leave an opening at its lower edge. Again, braces can help close the gaps.
Extra help needed to clamp the rubber to the hull.
Finally,
and even after some steaming, there was a snap and the rubber
broke right at the point where the sheer strake did all those years
ago. I am reluctant to start over because of the screw holes already in
the sheer. You never get them quite right a second time. So, I will try
to repair the break.
The gunwale rubber snaps at the sharpest part of the curve.
Back to the bloody epoxy
again! The gunwale rubber snaps at the sharpest part of the curve.
Clamping the repair.
While
waiting for that to set I have given some thought to the look of the
rubber. It is, as I feared, excessively unattractive because of its
width. But I do not think a semicircular profile would help either. It
is just too plod-like. I will chamfer the top and bottom edges of the
strip, as I did for the lower strake, and leave the middle unshaped.
But to give it a touch of pizzazz I am going to fill the screw holes
with black plugs. Some Wenge should do the job nicely, and that will
complement an ebony cap which I will put on top of the stem to cover
its end grain.
The repair looks good. Now it will have to stand up to bending again. This time the whole area will be soaked in boiling water for half an hour before attempting the attachment. As long as the temperature is not high enough to soften the epoxy, which I believe it is not, there should be a good chance of getting it to take the curve.
This repair should be just as strong as a scarf joint.
The repair looks good. Now it will have to stand up to bending again. This time the whole area will be soaked in boiling water for half an hour before attempting the attachment. As long as the temperature is not high enough to soften the epoxy, which I believe it is not, there should be a good chance of getting it to take the curve.
This repair should be just as strong as a scarf joint.
To
soak
the piece I wrap a towel around it and pour boiling water over it
repeatedly. Only one section of it needs to be steamed in this way,
because the rest of it is a fairly straight run.
But, the result is the same again. The snap is straight through the original, splitting apart the epoxy join. So much for its superior strength.
The
thought occurs to me that part of the problem here might be the
weakening effect of the 1/2" bores necessary to take the 12 gauge screw
heads. Cutting the leading end on another piece of wood, and fixing it
with a single screw to the bow, I am able to bend it around the hull
without any streaming at all. So I think the solution here is to leave
it attached like this for a week to train the bend into it, and then
try screwing it in again. In the meantime I can get on with other small
tasks.
The next piece takes the bend with no trouble.
The next piece takes the bend with no trouble.
After
successfully
leaving the stick clamped to the hull for a week, however,
the same thing happened when I started drilling for the screws. This
topic has now earned itself a chapter in the Problems
page.
The third split.
The third split.
A
new
straight grained piece of Merbau behaved much better, and when I
eventually got sick of dodging the clamps which were holding the
forward rubbers in their training positions, I bit the bullet again,
with much more success. Both forward sticks took the bend without any
of those groans and creaks which accompanied the previous ones. They
are now attached to the sheer, but will need to be removed again for
scarfing with the next pieces.
The successfully attached forward rubbers.
The successfully attached forward rubbers.
Success
at
this stage, however, will not necessarily follow through. The
question now is whether to scarf the full length rubber together, or to
continue with the unjoined strips. The rear segment will probably give
me even more grief than the forward one has. It has to bend down to
follow the slope of the deck, and it has 36 mm. of stock to resist the
bend. If I go ahead and scarf now, and then the rear section breaks I
am stuck again, but if I try to bend the rear section by itself it will
make for a difficult scarfing procedure to get the components to join
up later, while still employing the existing screw holes.
The decision is to proceed with the unjoined segments, and to try to bend the rear section to shape next, preferably without screwing it onto the hull. If that is successful I will scarf the full length and screw it all on. The experiment satisfied me that the rear member would take the strain, so it is back to the scarfing jig again.
The decision is to proceed with the unjoined segments, and to try to bend the rear section to shape next, preferably without screwing it onto the hull. If that is successful I will scarf the full length and screw it all on. The experiment satisfied me that the rear member would take the strain, so it is back to the scarfing jig again.
This
time
there is no trouble getting them to take the curves, and the still
unshaped rubbers are screwed into place without any adhesive. Before I
attempt to glue them on I want to mask off the top strake, and get some
finish onto the covering boards so that the squeeze out does not stain
them permanently. The previous idea of putting a small inlay into the
gap between the covering boards and the rubbers is not necessary. The
great majority of the length of these joins shows no gap at all. In the
area of the scarf joints there are some which have opened up a little,
but they can be obscured by slipping some fine shavings of Merbau into
them when the glue is wet, and then sanding them off later.
The starboard rubber in the process of being screwed on, and the port already done.
While I am getting the covering boards ready for a coat of finish, I may as well do the rest of the deck in preparation for the caulking, so the final attachment of the gunwale will have to wait until then.
The starboard rubber in the process of being screwed on, and the port already done.
While I am getting the covering boards ready for a coat of finish, I may as well do the rest of the deck in preparation for the caulking, so the final attachment of the gunwale will have to wait until then.
